The Big Thirst: How Is Japan’s Fukushima Nuclear Plant Making "Radioactive" Water?

In this installment, "The Big Thirst" author and Fast Company writer explores how water, which technically can't be made radioactive, could be the least threatening byproduct of the hobbled Fukushima plant.

FACT: Nothing is thirstier than nuclear power plants. They use water deep inside the reactor core, and they use rivers of water for cooling.

 One typical U.S. nuclear power plant uses 30 million gallons of cooling water an hour. The whole city of New York uses 46 million gallons of water an hour, so a single nuclear power plant needs water flow that would support a city of about 5 million people (about as many as live in the Washington metro area). And the U.S. has 104 nuclear power plants—more than any other country, a quarter of all plants worldwide.

Today, the day after the one-month anniversary of the Japanese earthquake and tsunami, the thousands of gallons of radioactive water that have been leaking from Japan's Fukushima Dai-ichi nuclear plant and back into the Pacific Ocean urgently remind us of the water required for nuclear power. Water is one of the hidden, rarely discussed environmental costs of nuclear power plants. Water, in the form of steam, is typically what's used to turn the heat from the nuclear reaction into energy—with a turbine. But nuclear power plants also use water to help shield the reactor core from the rest of the facility, and the rest of the world.

That's what makes the "radioactive water" coming from Fukushima so puzzling. Water cannot be made radioactive. It simply won't absorb the waves of neutrons being put out by the reactor cores. It's in part why nuclear power plants use water right inside the reactor. It's why "spent" nuclear fuel is typically stored immersed in pools of water. The water keeps the fuel cool, but equally valuable, it is a great radiation shield.

So in what way is the water pouring from Fukushima into the Pacific radioactive?

The water accumulating around the Fukushima reactors, and then leaking into the ocean, is dirty, and it's the debris that is radioactive. The debris may be ordinary dirt, material from the buildings after the explosions, or the failing metal and concrete structure of the reactors themselves.

But since the water itself is not radioactive, it can be cleaned. Water, in fact, is one of the few substances that can be de-radiated. The radiation can be taken from it by simply filtering—basic filtering of the dirt, and more sophisticated filtering, using something called ion-exchange beds, to get both suspended dirty and dissolved substances out.

It's a bit more complicated than running the stuff from your tap through a Brita filter. But it's also not that different. In fact, every nuclear reactor does precisely that kind of filtering all the time—nuclear plants routinely super-clean their reactor water, precisely to prevent any debris from starting to circulate (corrosion from pipes or valves, for instance) that would quickly become radioactive.

The Fukushima water can be cleaned as simply as allowing it to evaporate, in open ponds. What will be left, of course, will be a layer of debris, which will be radioactive. But that material was radioactive to start—before it contaminated the water. Disposing of radioactive debris is mostly a matter of security and time—but solids don't migrate the way water carrying radioactivity does.

The Fukushima disaster suddenly reminds us how important water is to nuclear power, but all the major ways we generate electricity require huge volumes of water.

In fact, electricity and water are intimately linked in a way we typically ignore.

In the United States, the single largest use of water is to run electric power plants: 49% of the water used in the U.S. goes to generate electricity. U.S. power plants use seven times as much water each day as homes. And the largest single users of electricity? The nation's water treatment plants and water pumping stations.

Adapted from The Big Thirst: The Secret Life and Turbulent Future of Water, to be published in April by Free Press / Simon & Schuster. © 2011, Charles Fishman.

Read the feature from Fast Company's April issue.

Read more from The Big Thirst on

Read more coverage of the disaster in Japan.

[Front page image: Flickr user AmyZZZ1; Top image: Flickr user daveeza]

Add New Comment


  • Fanandala

    "A nuclear power station "uses" 30 Million gallons of water an hour" ... warms it up by a couple of degrees, and releases it back into the river lake or ocean from where it came from originally.  Not necessarily potable water. A coal fired power station of the same capacity would need just as much water. 

  • Ace Hoffman


    Tritium doesn't usually come from normal hydrogen in water, which would have to first become deuterium and then tritium. Nor does most of it come from the small amount of deuterium already in LWRs (Light Water Reactors) although in Canadian CANDU style reactors, I believe most it does come from the heavy water. After all, what's heavy about it is the deuterium. Tritium usually comes from lithium-6 which picks up a neutron and becomes lithium-7, which then splits, and one half of it is tritium, and the other half is (stable) helium.

    As to "borrowing" the water as one commentator describes it (and thermally shocking the local flora and fauna both when they turn it on and when then suddenly shut it off), ALL reactors leak primary coolant into secondary coolant loops, and secondary coolant loops leak into tertiary coolant loops (for PWRs (Pressurized Water Reactors. BWRs (Boiling Water Reactors) have only two coolant loops, not three)).

    All reactors leak. Those with cooling towers turn millions of gallons of water into steam every day. Some of the steam is HTO, not H2O. Tritium has been found under dozens of U. S. reactors. It's probably under all of them but perhaps they haven't looked yet for most of them. But they leak everything, vent noble gases constantly, and have accidents that should be included in any calculation of what the "average" plant puts out, because accidents happen and will continue to happen if we keep using this very flawed technology. Tritium is usually "removed" by evaporation...

    You cannot filter out tritium from normal water with normal methods available to average citizens. Nuclear scientists with large budgets can do it.

    Also, I would say that a fairly substantial amount of radioactive material in these pools would also be lofted into the air during the process of evaporation. You would want to filter the evaporated air.

    But, and lastly, HEPA filters and ion exchanges and so forth only do a partial job. And filters break down very quickly in radioactive environments. And filters need to be checked regularly and changed frequently to work at all. See what happened at Davis-Bese in 2002 if you think the nuclear industry can be relied on to change the filters regularly, or even check them!

    But yeah, nuclear reactors use (and poison) and awful lot of water, and as water becomes more valuable and we all see what nuclear power can do to billions of gallons of water a day (poison it for years to come...) perhaps we'll all finally take a proper look at -- and reject -- this technology.

    At least I hope so!

  • Atomik Rabbit

    “One typical U.S. nuclear power plant uses 30 million gallons of cooling water an hour”

    John Wheeler has pointed out – this could be misinterpreted, depending on what your understanding of “uses” is. The water is not “consumed” and now unavailable.

    As Fishman touches on briefly at the end of the article, all steam thermal power plants, whether coal, gas, geothermal, concentrating solar, or nuclear “use” water in this manner to condense the turbine-generator’s steam, recycling it to the environment a few degrees warmer. The amount of water “used” depends generally on the amount of steam being condensed, which is roughly proportional to the amount of energy generated. So larger generating plants, like most nukes, will “use” more water.

    The Palo Verde Nuclear Generating Station west of Phoenix, however, produces 3200 megawatts of electricity while evaporating water from the treated sewage of several nearby municipalities to meet its cooling needs.

    Brian - As far as the explanation of the Fukushima water radioactivity, of course every nuclear physicist knows that a water hydrogen has a certain small cross-section for absorption of a neutron to form deuterium. This is why heavy (deuterated) water is a better neutron moderator than light (regular) water – the nucleus is already relatively “saturated” with neutrons and is unable to parasitically absorb another (lets not further complicate this by bringing up tritium).

    I think Fishman did a fair job in explaining this in layman’s terms and pointed out that the cleanup will be technically challenging but doable, and he avoided most of the fear-mongering we sometimes see.

    The bottom line is, after the worst regional earthquake and tsunami in a millennium, a set of 40-year old nuclear power plants, due to the engineering concept of defense-in-depth, can still boast no one killed or even seriously injured from the awesome source of energy they have harnessed. No other electrical generation source is even close when it comes to safety, especially on a per-MW-generated basis, and I include wind and solar in that:

  • John Wheeler

    Saying that nuclear plants and other electricity generators "use" water is very misleading. It is more accurate to say power plants "borrow" the water for cooling and return it a few degrees warmer. The water flows in and then it flows out. After it leaves the power plant the water remains available for recreation, drinking, the environment, and other purposes.

    In addition, while Mr Fishman would lead us to believe the cooling water from rivers, lakes and the sea goes into nuclear reactors to become irradiated, that is not true. Except for rare accident conditions (like at Fukushima), cooling water that returns to the environment does not come into contact with radioactive parts of a nuclear power plant, and is not radioactive except for small amounts of naturally occurring radioactive materials already in the environment.

    John Wheeler
    Producer, This Week in Nuclear

  • brian grainger

    Do you ever fact check beofre you publish this stuff? Yes, there are radioactive verisons of water. the hydrogen in water can be irradiated to Deuterium or Tritium. So H20 becomes D2O.